1# SPDX-License-Identifier: GPL-2.0-only 2menu "Kernel hacking" 3 4menu "printk and dmesg options" 5 6config PRINTK_TIME 7 bool "Show timing information on printks" 8 depends on PRINTK 9 help 10 Selecting this option causes time stamps of the printk() 11 messages to be added to the output of the syslog() system 12 call and at the console. 13 14 The timestamp is always recorded internally, and exported 15 to /dev/kmsg. This flag just specifies if the timestamp should 16 be included, not that the timestamp is recorded. 17 18 The behavior is also controlled by the kernel command line 19 parameter printk.time=1. See Documentation/admin-guide/kernel-parameters.rst 20 21config PRINTK_CALLER 22 bool "Show caller information on printks" 23 depends on PRINTK 24 help 25 Selecting this option causes printk() to add a caller "thread id" (if 26 in task context) or a caller "processor id" (if not in task context) 27 to every message. 28 29 This option is intended for environments where multiple threads 30 concurrently call printk() for many times, for it is difficult to 31 interpret without knowing where these lines (or sometimes individual 32 line which was divided into multiple lines due to race) came from. 33 34 Since toggling after boot makes the code racy, currently there is 35 no option to enable/disable at the kernel command line parameter or 36 sysfs interface. 37 38config CONSOLE_LOGLEVEL_DEFAULT 39 int "Default console loglevel (1-15)" 40 range 1 15 41 default "7" 42 help 43 Default loglevel to determine what will be printed on the console. 44 45 Setting a default here is equivalent to passing in loglevel=<x> in 46 the kernel bootargs. loglevel=<x> continues to override whatever 47 value is specified here as well. 48 49 Note: This does not affect the log level of un-prefixed printk() 50 usage in the kernel. That is controlled by the MESSAGE_LOGLEVEL_DEFAULT 51 option. 52 53config CONSOLE_LOGLEVEL_QUIET 54 int "quiet console loglevel (1-15)" 55 range 1 15 56 default "4" 57 help 58 loglevel to use when "quiet" is passed on the kernel commandline. 59 60 When "quiet" is passed on the kernel commandline this loglevel 61 will be used as the loglevel. IOW passing "quiet" will be the 62 equivalent of passing "loglevel=<CONSOLE_LOGLEVEL_QUIET>" 63 64config MESSAGE_LOGLEVEL_DEFAULT 65 int "Default message log level (1-7)" 66 range 1 7 67 default "4" 68 help 69 Default log level for printk statements with no specified priority. 70 71 This was hard-coded to KERN_WARNING since at least 2.6.10 but folks 72 that are auditing their logs closely may want to set it to a lower 73 priority. 74 75 Note: This does not affect what message level gets printed on the console 76 by default. To change that, use loglevel=<x> in the kernel bootargs, 77 or pick a different CONSOLE_LOGLEVEL_DEFAULT configuration value. 78 79config BOOT_PRINTK_DELAY 80 bool "Delay each boot printk message by N milliseconds" 81 depends on DEBUG_KERNEL && PRINTK && GENERIC_CALIBRATE_DELAY 82 help 83 This build option allows you to read kernel boot messages 84 by inserting a short delay after each one. The delay is 85 specified in milliseconds on the kernel command line, 86 using "boot_delay=N". 87 88 It is likely that you would also need to use "lpj=M" to preset 89 the "loops per jiffie" value. 90 See a previous boot log for the "lpj" value to use for your 91 system, and then set "lpj=M" before setting "boot_delay=N". 92 NOTE: Using this option may adversely affect SMP systems. 93 I.e., processors other than the first one may not boot up. 94 BOOT_PRINTK_DELAY also may cause LOCKUP_DETECTOR to detect 95 what it believes to be lockup conditions. 96 97config DYNAMIC_DEBUG 98 bool "Enable dynamic printk() support" 99 default n 100 depends on PRINTK 101 depends on (DEBUG_FS || PROC_FS) 102 help 103 104 Compiles debug level messages into the kernel, which would not 105 otherwise be available at runtime. These messages can then be 106 enabled/disabled based on various levels of scope - per source file, 107 function, module, format string, and line number. This mechanism 108 implicitly compiles in all pr_debug() and dev_dbg() calls, which 109 enlarges the kernel text size by about 2%. 110 111 If a source file is compiled with DEBUG flag set, any 112 pr_debug() calls in it are enabled by default, but can be 113 disabled at runtime as below. Note that DEBUG flag is 114 turned on by many CONFIG_*DEBUG* options. 115 116 Usage: 117 118 Dynamic debugging is controlled via the 'dynamic_debug/control' file, 119 which is contained in the 'debugfs' filesystem or procfs. 120 Thus, the debugfs or procfs filesystem must first be mounted before 121 making use of this feature. 122 We refer the control file as: <debugfs>/dynamic_debug/control. This 123 file contains a list of the debug statements that can be enabled. The 124 format for each line of the file is: 125 126 filename:lineno [module]function flags format 127 128 filename : source file of the debug statement 129 lineno : line number of the debug statement 130 module : module that contains the debug statement 131 function : function that contains the debug statement 132 flags : '=p' means the line is turned 'on' for printing 133 format : the format used for the debug statement 134 135 From a live system: 136 137 nullarbor:~ # cat <debugfs>/dynamic_debug/control 138 # filename:lineno [module]function flags format 139 fs/aio.c:222 [aio]__put_ioctx =_ "__put_ioctx:\040freeing\040%p\012" 140 fs/aio.c:248 [aio]ioctx_alloc =_ "ENOMEM:\040nr_events\040too\040high\012" 141 fs/aio.c:1770 [aio]sys_io_cancel =_ "calling\040cancel\012" 142 143 Example usage: 144 145 // enable the message at line 1603 of file svcsock.c 146 nullarbor:~ # echo -n 'file svcsock.c line 1603 +p' > 147 <debugfs>/dynamic_debug/control 148 149 // enable all the messages in file svcsock.c 150 nullarbor:~ # echo -n 'file svcsock.c +p' > 151 <debugfs>/dynamic_debug/control 152 153 // enable all the messages in the NFS server module 154 nullarbor:~ # echo -n 'module nfsd +p' > 155 <debugfs>/dynamic_debug/control 156 157 // enable all 12 messages in the function svc_process() 158 nullarbor:~ # echo -n 'func svc_process +p' > 159 <debugfs>/dynamic_debug/control 160 161 // disable all 12 messages in the function svc_process() 162 nullarbor:~ # echo -n 'func svc_process -p' > 163 <debugfs>/dynamic_debug/control 164 165 See Documentation/admin-guide/dynamic-debug-howto.rst for additional 166 information. 167 168config SYMBOLIC_ERRNAME 169 bool "Support symbolic error names in printf" 170 default y if PRINTK 171 help 172 If you say Y here, the kernel's printf implementation will 173 be able to print symbolic error names such as ENOSPC instead 174 of the number 28. It makes the kernel image slightly larger 175 (about 3KB), but can make the kernel logs easier to read. 176 177config DEBUG_BUGVERBOSE 178 bool "Verbose BUG() reporting (adds 70K)" if DEBUG_KERNEL && EXPERT 179 depends on BUG && (GENERIC_BUG || HAVE_DEBUG_BUGVERBOSE) 180 default y 181 help 182 Say Y here to make BUG() panics output the file name and line number 183 of the BUG call as well as the EIP and oops trace. This aids 184 debugging but costs about 70-100K of memory. 185 186endmenu # "printk and dmesg options" 187 188menu "Compile-time checks and compiler options" 189 190config DEBUG_INFO 191 bool "Compile the kernel with debug info" 192 depends on DEBUG_KERNEL && !COMPILE_TEST 193 help 194 If you say Y here the resulting kernel image will include 195 debugging info resulting in a larger kernel image. 196 This adds debug symbols to the kernel and modules (gcc -g), and 197 is needed if you intend to use kernel crashdump or binary object 198 tools like crash, kgdb, LKCD, gdb, etc on the kernel. 199 Say Y here only if you plan to debug the kernel. 200 201 If unsure, say N. 202 203config DEBUG_INFO_REDUCED 204 bool "Reduce debugging information" 205 depends on DEBUG_INFO 206 help 207 If you say Y here gcc is instructed to generate less debugging 208 information for structure types. This means that tools that 209 need full debugging information (like kgdb or systemtap) won't 210 be happy. But if you merely need debugging information to 211 resolve line numbers there is no loss. Advantage is that 212 build directory object sizes shrink dramatically over a full 213 DEBUG_INFO build and compile times are reduced too. 214 Only works with newer gcc versions. 215 216config DEBUG_INFO_SPLIT 217 bool "Produce split debuginfo in .dwo files" 218 depends on DEBUG_INFO 219 depends on $(cc-option,-gsplit-dwarf) 220 help 221 Generate debug info into separate .dwo files. This significantly 222 reduces the build directory size for builds with DEBUG_INFO, 223 because it stores the information only once on disk in .dwo 224 files instead of multiple times in object files and executables. 225 In addition the debug information is also compressed. 226 227 Requires recent gcc (4.7+) and recent gdb/binutils. 228 Any tool that packages or reads debug information would need 229 to know about the .dwo files and include them. 230 Incompatible with older versions of ccache. 231 232config DEBUG_INFO_DWARF4 233 bool "Generate dwarf4 debuginfo" 234 depends on DEBUG_INFO 235 depends on $(cc-option,-gdwarf-4) 236 help 237 Generate dwarf4 debug info. This requires recent versions 238 of gcc and gdb. It makes the debug information larger. 239 But it significantly improves the success of resolving 240 variables in gdb on optimized code. 241 242config DEBUG_INFO_BTF 243 bool "Generate BTF typeinfo" 244 depends on DEBUG_INFO 245 depends on !DEBUG_INFO_SPLIT && !DEBUG_INFO_REDUCED 246 depends on !GCC_PLUGIN_RANDSTRUCT || COMPILE_TEST 247 help 248 Generate deduplicated BTF type information from DWARF debug info. 249 Turning this on expects presence of pahole tool, which will convert 250 DWARF type info into equivalent deduplicated BTF type info. 251 252config GDB_SCRIPTS 253 bool "Provide GDB scripts for kernel debugging" 254 depends on DEBUG_INFO 255 help 256 This creates the required links to GDB helper scripts in the 257 build directory. If you load vmlinux into gdb, the helper 258 scripts will be automatically imported by gdb as well, and 259 additional functions are available to analyze a Linux kernel 260 instance. See Documentation/dev-tools/gdb-kernel-debugging.rst 261 for further details. 262 263config ENABLE_MUST_CHECK 264 bool "Enable __must_check logic" 265 default y 266 help 267 Enable the __must_check logic in the kernel build. Disable this to 268 suppress the "warning: ignoring return value of 'foo', declared with 269 attribute warn_unused_result" messages. 270 271config FRAME_WARN 272 int "Warn for stack frames larger than" 273 range 0 8192 274 default 2048 if GCC_PLUGIN_LATENT_ENTROPY 275 default 1280 if (!64BIT && PARISC) 276 default 1024 if (!64BIT && !PARISC) 277 default 2048 if 64BIT 278 help 279 Tell gcc to warn at build time for stack frames larger than this. 280 Setting this too low will cause a lot of warnings. 281 Setting it to 0 disables the warning. 282 283config STRIP_ASM_SYMS 284 bool "Strip assembler-generated symbols during link" 285 default n 286 help 287 Strip internal assembler-generated symbols during a link (symbols 288 that look like '.Lxxx') so they don't pollute the output of 289 get_wchan() and suchlike. 290 291config READABLE_ASM 292 bool "Generate readable assembler code" 293 depends on DEBUG_KERNEL 294 help 295 Disable some compiler optimizations that tend to generate human unreadable 296 assembler output. This may make the kernel slightly slower, but it helps 297 to keep kernel developers who have to stare a lot at assembler listings 298 sane. 299 300config HEADERS_INSTALL 301 bool "Install uapi headers to usr/include" 302 depends on !UML 303 help 304 This option will install uapi headers (headers exported to user-space) 305 into the usr/include directory for use during the kernel build. 306 This is unneeded for building the kernel itself, but needed for some 307 user-space program samples. It is also needed by some features such 308 as uapi header sanity checks. 309 310config DEBUG_SECTION_MISMATCH 311 bool "Enable full Section mismatch analysis" 312 help 313 The section mismatch analysis checks if there are illegal 314 references from one section to another section. 315 During linktime or runtime, some sections are dropped; 316 any use of code/data previously in these sections would 317 most likely result in an oops. 318 In the code, functions and variables are annotated with 319 __init,, etc. (see the full list in include/linux/init.h), 320 which results in the code/data being placed in specific sections. 321 The section mismatch analysis is always performed after a full 322 kernel build, and enabling this option causes the following 323 additional step to occur: 324 - Add the option -fno-inline-functions-called-once to gcc commands. 325 When inlining a function annotated with __init in a non-init 326 function, we would lose the section information and thus 327 the analysis would not catch the illegal reference. 328 This option tells gcc to inline less (but it does result in 329 a larger kernel). 330 331config SECTION_MISMATCH_WARN_ONLY 332 bool "Make section mismatch errors non-fatal" 333 default y 334 help 335 If you say N here, the build process will fail if there are any 336 section mismatch, instead of just throwing warnings. 337 338 If unsure, say Y. 339 340# 341# Select this config option from the architecture Kconfig, if it 342# is preferred to always offer frame pointers as a config 343# option on the architecture (regardless of KERNEL_DEBUG): 344# 345config ARCH_WANT_FRAME_POINTERS 346 bool 347 348config FRAME_POINTER 349 bool "Compile the kernel with frame pointers" 350 depends on DEBUG_KERNEL && (M68K || UML || SUPERH) || ARCH_WANT_FRAME_POINTERS 351 default y if (DEBUG_INFO && UML) || ARCH_WANT_FRAME_POINTERS 352 help 353 If you say Y here the resulting kernel image will be slightly 354 larger and slower, but it gives very useful debugging information 355 in case of kernel bugs. (precise oopses/stacktraces/warnings) 356 357config STACK_VALIDATION 358 bool "Compile-time stack metadata validation" 359 depends on HAVE_STACK_VALIDATION 360 default n 361 help 362 Add compile-time checks to validate stack metadata, including frame 363 pointers (if CONFIG_FRAME_POINTER is enabled). This helps ensure 364 that runtime stack traces are more reliable. 365 366 This is also a prerequisite for generation of ORC unwind data, which 367 is needed for CONFIG_UNWINDER_ORC. 368 369 For more information, see 370 tools/objtool/Documentation/stack-validation.txt. 371 372config VMLINUX_VALIDATION 373 bool 374 depends on STACK_VALIDATION && DEBUG_ENTRY && !PARAVIRT 375 default y 376 377config DEBUG_FORCE_WEAK_PER_CPU 378 bool "Force weak per-cpu definitions" 379 depends on DEBUG_KERNEL 380 help 381 s390 and alpha require percpu variables in modules to be 382 defined weak to work around addressing range issue which 383 puts the following two restrictions on percpu variable 384 definitions. 385 386 1. percpu symbols must be unique whether static or not 387 2. percpu variables can't be defined inside a function 388 389 To ensure that generic code follows the above rules, this 390 option forces all percpu variables to be defined as weak. 391 392endmenu # "Compiler options" 393 394menu "Generic Kernel Debugging Instruments" 395 396config MAGIC_SYSRQ 397 bool "Magic SysRq key" 398 depends on !UML 399 help 400 If you say Y here, you will have some control over the system even 401 if the system crashes for example during kernel debugging (e.g., you 402 will be able to flush the buffer cache to disk, reboot the system 403 immediately or dump some status information). This is accomplished 404 by pressing various keys while holding SysRq (Alt+PrintScreen). It 405 also works on a serial console (on PC hardware at least), if you 406 send a BREAK and then within 5 seconds a command keypress. The 407 keys are documented in <file:Documentation/admin-guide/sysrq.rst>. 408 Don't say Y unless you really know what this hack does. 409 410config MAGIC_SYSRQ_DEFAULT_ENABLE 411 hex "Enable magic SysRq key functions by default" 412 depends on MAGIC_SYSRQ 413 default 0x1 414 help 415 Specifies which SysRq key functions are enabled by default. 416 This may be set to 1 or 0 to enable or disable them all, or 417 to a bitmask as described in Documentation/admin-guide/sysrq.rst. 418 419config MAGIC_SYSRQ_SERIAL 420 bool "Enable magic SysRq key over serial" 421 depends on MAGIC_SYSRQ 422 default y 423 help 424 Many embedded boards have a disconnected TTL level serial which can 425 generate some garbage that can lead to spurious false sysrq detects. 426 This option allows you to decide whether you want to enable the 427 magic SysRq key. 428 429config MAGIC_SYSRQ_SERIAL_SEQUENCE 430 string "Char sequence that enables magic SysRq over serial" 431 depends on MAGIC_SYSRQ_SERIAL 432 default "" 433 help 434 Specifies a sequence of characters that can follow BREAK to enable 435 SysRq on a serial console. 436 437 If unsure, leave an empty string and the option will not be enabled. 438 439config DEBUG_FS 440 bool "Debug Filesystem" 441 help 442 debugfs is a virtual file system that kernel developers use to put 443 debugging files into. Enable this option to be able to read and 444 write to these files. 445 446 For detailed documentation on the debugfs API, see 447 Documentation/filesystems/. 448 449 If unsure, say N. 450 451source "lib/Kconfig.kgdb" 452 453source "lib/Kconfig.ubsan" 454 455endmenu 456 457config DEBUG_KERNEL 458 bool "Kernel debugging" 459 help 460 Say Y here if you are developing drivers or trying to debug and 461 identify kernel problems. 462 463config DEBUG_MISC 464 bool "Miscellaneous debug code" 465 default DEBUG_KERNEL 466 depends on DEBUG_KERNEL 467 help 468 Say Y here if you need to enable miscellaneous debug code that should 469 be under a more specific debug option but isn't. 470 471 472menu "Memory Debugging" 473 474source "mm/Kconfig.debug" 475 476config DEBUG_OBJECTS 477 bool "Debug object operations" 478 depends on DEBUG_KERNEL 479 help 480 If you say Y here, additional code will be inserted into the 481 kernel to track the life time of various objects and validate 482 the operations on those objects. 483 484config DEBUG_OBJECTS_SELFTEST 485 bool "Debug objects selftest" 486 depends on DEBUG_OBJECTS 487 help 488 This enables the selftest of the object debug code. 489 490config DEBUG_OBJECTS_FREE 491 bool "Debug objects in freed memory" 492 depends on DEBUG_OBJECTS 493 help 494 This enables checks whether a k/v free operation frees an area 495 which contains an object which has not been deactivated 496 properly. This can make kmalloc/kfree-intensive workloads 497 much slower. 498 499config DEBUG_OBJECTS_TIMERS 500 bool "Debug timer objects" 501 depends on DEBUG_OBJECTS 502 help 503 If you say Y here, additional code will be inserted into the 504 timer routines to track the life time of timer objects and 505 validate the timer operations. 506 507config DEBUG_OBJECTS_WORK 508 bool "Debug work objects" 509 depends on DEBUG_OBJECTS 510 help 511 If you say Y here, additional code will be inserted into the 512 work queue routines to track the life time of work objects and 513 validate the work operations. 514 515config DEBUG_OBJECTS_RCU_HEAD 516 bool "Debug RCU callbacks objects" 517 depends on DEBUG_OBJECTS 518 help 519 Enable this to turn on debugging of RCU list heads (call_rcu() usage). 520 521config DEBUG_OBJECTS_PERCPU_COUNTER 522 bool "Debug percpu counter objects" 523 depends on DEBUG_OBJECTS 524 help 525 If you say Y here, additional code will be inserted into the 526 percpu counter routines to track the life time of percpu counter 527 objects and validate the percpu counter operations. 528 529config DEBUG_OBJECTS_ENABLE_DEFAULT 530 int "debug_objects bootup default value (0-1)" 531 range 0 1 532 default "1" 533 depends on DEBUG_OBJECTS 534 help 535 Debug objects boot parameter default value 536 537config DEBUG_SLAB 538 bool "Debug slab memory allocations" 539 depends on DEBUG_KERNEL && SLAB 540 help 541 Say Y here to have the kernel do limited verification on memory 542 allocation as well as poisoning memory on free to catch use of freed 543 memory. This can make kmalloc/kfree-intensive workloads much slower. 544 545config SLUB_DEBUG_ON 546 bool "SLUB debugging on by default" 547 depends on SLUB && SLUB_DEBUG 548 default n 549 help 550 Boot with debugging on by default. SLUB boots by default with 551 the runtime debug capabilities switched off. Enabling this is 552 equivalent to specifying the "slub_debug" parameter on boot. 553 There is no support for more fine grained debug control like 554 possible with slub_debug=xxx. SLUB debugging may be switched 555 off in a kernel built with CONFIG_SLUB_DEBUG_ON by specifying 556 "slub_debug=-". 557 558config SLUB_STATS 559 default n 560 bool "Enable SLUB performance statistics" 561 depends on SLUB && SYSFS 562 help 563 SLUB statistics are useful to debug SLUBs allocation behavior in 564 order find ways to optimize the allocator. This should never be 565 enabled for production use since keeping statistics slows down 566 the allocator by a few percentage points. The slabinfo command 567 supports the determination of the most active slabs to figure 568 out which slabs are relevant to a particular load. 569 Try running: slabinfo -DA 570 571config HAVE_DEBUG_KMEMLEAK 572 bool 573 574config DEBUG_KMEMLEAK 575 bool "Kernel memory leak detector" 576 depends on DEBUG_KERNEL && HAVE_DEBUG_KMEMLEAK 577 select DEBUG_FS 578 select STACKTRACE if STACKTRACE_SUPPORT 579 select KALLSYMS 580 select CRC32 581 help 582 Say Y here if you want to enable the memory leak 583 detector. The memory allocation/freeing is traced in a way 584 similar to the Boehm's conservative garbage collector, the 585 difference being that the orphan objects are not freed but 586 only shown in /sys/kernel/debug/kmemleak. Enabling this 587 feature will introduce an overhead to memory 588 allocations. See Documentation/dev-tools/kmemleak.rst for more 589 details. 590 591 Enabling DEBUG_SLAB or SLUB_DEBUG may increase the chances 592 of finding leaks due to the slab objects poisoning. 593 594 In order to access the kmemleak file, debugfs needs to be 595 mounted (usually at /sys/kernel/debug). 596 597config DEBUG_KMEMLEAK_MEM_POOL_SIZE 598 int "Kmemleak memory pool size" 599 depends on DEBUG_KMEMLEAK 600 range 200 1000000 601 default 16000 602 help 603 Kmemleak must track all the memory allocations to avoid 604 reporting false positives. Since memory may be allocated or 605 freed before kmemleak is fully initialised, use a static pool 606 of metadata objects to track such callbacks. After kmemleak is 607 fully initialised, this memory pool acts as an emergency one 608 if slab allocations fail. 609 610config DEBUG_KMEMLEAK_TEST 611 tristate "Simple test for the kernel memory leak detector" 612 depends on DEBUG_KMEMLEAK && m 613 help 614 This option enables a module that explicitly leaks memory. 615 616 If unsure, say N. 617 618config DEBUG_KMEMLEAK_DEFAULT_OFF 619 bool "Default kmemleak to off" 620 depends on DEBUG_KMEMLEAK 621 help 622 Say Y here to disable kmemleak by default. It can then be enabled 623 on the command line via kmemleak=on. 624 625config DEBUG_KMEMLEAK_AUTO_SCAN 626 bool "Enable kmemleak auto scan thread on boot up" 627 default y 628 depends on DEBUG_KMEMLEAK 629 help 630 Depending on the cpu, kmemleak scan may be cpu intensive and can 631 stall user tasks at times. This option enables/disables automatic 632 kmemleak scan at boot up. 633 634 Say N here to disable kmemleak auto scan thread to stop automatic 635 scanning. Disabling this option disables automatic reporting of 636 memory leaks. 637 638 If unsure, say Y. 639 640config DEBUG_STACK_USAGE 641 bool "Stack utilization instrumentation" 642 depends on DEBUG_KERNEL && !IA64 643 help 644 Enables the display of the minimum amount of free stack which each 645 task has ever had available in the sysrq-T and sysrq-P debug output. 646 647 This option will slow down process creation somewhat. 648 649config SCHED_STACK_END_CHECK 650 bool "Detect stack corruption on calls to schedule()" 651 depends on DEBUG_KERNEL 652 default n 653 help 654 This option checks for a stack overrun on calls to schedule(). 655 If the stack end location is found to be over written always panic as 656 the content of the corrupted region can no longer be trusted. 657 This is to ensure no erroneous behaviour occurs which could result in 658 data corruption or a sporadic crash at a later stage once the region 659 is examined. The runtime overhead introduced is minimal. 660 661config DEBUG_VM 662 bool "Debug VM" 663 depends on DEBUG_KERNEL 664 help 665 Enable this to turn on extended checks in the virtual-memory system 666 that may impact performance. 667 668 If unsure, say N. 669 670config DEBUG_VM_VMACACHE 671 bool "Debug VMA caching" 672 depends on DEBUG_VM 673 help 674 Enable this to turn on VMA caching debug information. Doing so 675 can cause significant overhead, so only enable it in non-production 676 environments. 677 678 If unsure, say N. 679 680config DEBUG_VM_RB 681 bool "Debug VM red-black trees" 682 depends on DEBUG_VM 683 help 684 Enable VM red-black tree debugging information and extra validations. 685 686 If unsure, say N. 687 688config DEBUG_VM_PGFLAGS 689 bool "Debug page-flags operations" 690 depends on DEBUG_VM 691 help 692 Enables extra validation on page flags operations. 693 694 If unsure, say N. 695 696config ARCH_HAS_DEBUG_VIRTUAL 697 bool 698 699config DEBUG_VIRTUAL 700 bool "Debug VM translations" 701 depends on DEBUG_KERNEL && ARCH_HAS_DEBUG_VIRTUAL 702 help 703 Enable some costly sanity checks in virtual to page code. This can 704 catch mistakes with virt_to_page() and friends. 705 706 If unsure, say N. 707 708config DEBUG_NOMMU_REGIONS 709 bool "Debug the global anon/private NOMMU mapping region tree" 710 depends on DEBUG_KERNEL && !MMU 711 help 712 This option causes the global tree of anonymous and private mapping 713 regions to be regularly checked for invalid topology. 714 715config DEBUG_MEMORY_INIT 716 bool "Debug memory initialisation" if EXPERT 717 default !EXPERT 718 help 719 Enable this for additional checks during memory initialisation. 720 The sanity checks verify aspects of the VM such as the memory model 721 and other information provided by the architecture. Verbose 722 information will be printed at KERN_DEBUG loglevel depending 723 on the mminit_loglevel= command-line option. 724 725 If unsure, say Y 726 727config MEMORY_NOTIFIER_ERROR_INJECT 728 tristate "Memory hotplug notifier error injection module" 729 depends on MEMORY_HOTPLUG_SPARSE && NOTIFIER_ERROR_INJECTION 730 help 731 This option provides the ability to inject artificial errors to 732 memory hotplug notifier chain callbacks. It is controlled through 733 debugfs interface under /sys/kernel/debug/notifier-error-inject/memory 734 735 If the notifier call chain should be failed with some events 736 notified, write the error code to "actions/<notifier event>/error". 737 738 Example: Inject memory hotplug offline error (-12 == -ENOMEM) 739 740 # cd /sys/kernel/debug/notifier-error-inject/memory 741 # echo -12 > actions/MEM_GOING_OFFLINE/error 742 # echo offline > /sys/devices/system/memory/memoryXXX/state 743 bash: echo: write error: Cannot allocate memory 744 745 To compile this code as a module, choose M here: the module will 746 be called memory-notifier-error-inject. 747 748 If unsure, say N. 749 750config DEBUG_PER_CPU_MAPS 751 bool "Debug access to per_cpu maps" 752 depends on DEBUG_KERNEL 753 depends on SMP 754 help 755 Say Y to verify that the per_cpu map being accessed has 756 been set up. This adds a fair amount of code to kernel memory 757 and decreases performance. 758 759 Say N if unsure. 760 761config DEBUG_HIGHMEM 762 bool "Highmem debugging" 763 depends on DEBUG_KERNEL && HIGHMEM 764 help 765 This option enables additional error checking for high memory 766 systems. Disable for production systems. 767 768config HAVE_DEBUG_STACKOVERFLOW 769 bool 770 771config DEBUG_STACKOVERFLOW 772 bool "Check for stack overflows" 773 depends on DEBUG_KERNEL && HAVE_DEBUG_STACKOVERFLOW 774 ---help--- 775 Say Y here if you want to check for overflows of kernel, IRQ 776 and exception stacks (if your architecture uses them). This 777 option will show detailed messages if free stack space drops 778 below a certain limit. 779 780 These kinds of bugs usually occur when call-chains in the 781 kernel get too deep, especially when interrupts are 782 involved. 783 784 Use this in cases where you see apparently random memory 785 corruption, especially if it appears in 'struct thread_info' 786 787 If in doubt, say "N". 788 789source "lib/Kconfig.kasan" 790 791endmenu # "Memory Debugging" 792 793config DEBUG_SHIRQ 794 bool "Debug shared IRQ handlers" 795 depends on DEBUG_KERNEL 796 help 797 Enable this to generate a spurious interrupt as soon as a shared 798 interrupt handler is registered, and just before one is deregistered. 799 Drivers ought to be able to handle interrupts coming in at those 800 points; some don't and need to be caught. 801 802menu "Debug Oops, Lockups and Hangs" 803 804config PANIC_ON_OOPS 805 bool "Panic on Oops" 806 help 807 Say Y here to enable the kernel to panic when it oopses. This 808 has the same effect as setting oops=panic on the kernel command 809 line. 810 811 This feature is useful to ensure that the kernel does not do 812 anything erroneous after an oops which could result in data 813 corruption or other issues. 814 815 Say N if unsure. 816 817config PANIC_ON_OOPS_VALUE 818 int 819 range 0 1 820 default 0 if !PANIC_ON_OOPS 821 default 1 if PANIC_ON_OOPS 822 823config PANIC_TIMEOUT 824 int "panic timeout" 825 default 0 826 help 827 Set the timeout value (in seconds) until a reboot occurs when the 828 the kernel panics. If n = 0, then we wait forever. A timeout 829 value n > 0 will wait n seconds before rebooting, while a timeout 830 value n < 0 will reboot immediately. 831 832config LOCKUP_DETECTOR 833 bool 834 835config SOFTLOCKUP_DETECTOR 836 bool "Detect Soft Lockups" 837 depends on DEBUG_KERNEL && !S390 838 select LOCKUP_DETECTOR 839 help 840 Say Y here to enable the kernel to act as a watchdog to detect 841 soft lockups. 842 843 Softlockups are bugs that cause the kernel to loop in kernel 844 mode for more than 20 seconds, without giving other tasks a 845 chance to run. The current stack trace is displayed upon 846 detection and the system will stay locked up. 847 848config BOOTPARAM_SOFTLOCKUP_PANIC 849 bool "Panic (Reboot) On Soft Lockups" 850 depends on SOFTLOCKUP_DETECTOR 851 help 852 Say Y here to enable the kernel to panic on "soft lockups", 853 which are bugs that cause the kernel to loop in kernel 854 mode for more than 20 seconds (configurable using the watchdog_thresh 855 sysctl), without giving other tasks a chance to run. 856 857 The panic can be used in combination with panic_timeout, 858 to cause the system to reboot automatically after a 859 lockup has been detected. This feature is useful for 860 high-availability systems that have uptime guarantees and 861 where a lockup must be resolved ASAP. 862 863 Say N if unsure. 864 865config BOOTPARAM_SOFTLOCKUP_PANIC_VALUE 866 int 867 depends on SOFTLOCKUP_DETECTOR 868 range 0 1 869 default 0 if !BOOTPARAM_SOFTLOCKUP_PANIC 870 default 1 if BOOTPARAM_SOFTLOCKUP_PANIC 871 872config HARDLOCKUP_DETECTOR_PERF 873 bool 874 select SOFTLOCKUP_DETECTOR 875 876# 877# Enables a timestamp based low pass filter to compensate for perf based 878# hard lockup detection which runs too fast due to turbo modes. 879# 880config HARDLOCKUP_CHECK_TIMESTAMP 881 bool 882 883# 884# arch/ can define HAVE_HARDLOCKUP_DETECTOR_ARCH to provide their own hard 885# lockup detector rather than the perf based detector. 886# 887config HARDLOCKUP_DETECTOR 888 bool "Detect Hard Lockups" 889 depends on DEBUG_KERNEL && !S390 890 depends on HAVE_HARDLOCKUP_DETECTOR_PERF || HAVE_HARDLOCKUP_DETECTOR_ARCH 891 select LOCKUP_DETECTOR 892 select HARDLOCKUP_DETECTOR_PERF if HAVE_HARDLOCKUP_DETECTOR_PERF 893 select HARDLOCKUP_DETECTOR_ARCH if HAVE_HARDLOCKUP_DETECTOR_ARCH 894 help 895 Say Y here to enable the kernel to act as a watchdog to detect 896 hard lockups. 897 898 Hardlockups are bugs that cause the CPU to loop in kernel mode 899 for more than 10 seconds, without letting other interrupts have a 900 chance to run. The current stack trace is displayed upon detection 901 and the system will stay locked up. 902 903config BOOTPARAM_HARDLOCKUP_PANIC 904 bool "Panic (Reboot) On Hard Lockups" 905 depends on HARDLOCKUP_DETECTOR 906 help 907 Say Y here to enable the kernel to panic on "hard lockups", 908 which are bugs that cause the kernel to loop in kernel 909 mode with interrupts disabled for more than 10 seconds (configurable 910 using the watchdog_thresh sysctl). 911 912 Say N if unsure. 913 914config BOOTPARAM_HARDLOCKUP_PANIC_VALUE 915 int 916 depends on HARDLOCKUP_DETECTOR 917 range 0 1 918 default 0 if !BOOTPARAM_HARDLOCKUP_PANIC 919 default 1 if BOOTPARAM_HARDLOCKUP_PANIC 920 921config DETECT_HUNG_TASK 922 bool "Detect Hung Tasks" 923 depends on DEBUG_KERNEL 924 default SOFTLOCKUP_DETECTOR 925 help 926 Say Y here to enable the kernel to detect "hung tasks", 927 which are bugs that cause the task to be stuck in 928 uninterruptible "D" state indefinitely. 929 930 When a hung task is detected, the kernel will print the 931 current stack trace (which you should report), but the 932 task will stay in uninterruptible state. If lockdep is 933 enabled then all held locks will also be reported. This 934 feature has negligible overhead. 935 936config DEFAULT_HUNG_TASK_TIMEOUT 937 int "Default timeout for hung task detection (in seconds)" 938 depends on DETECT_HUNG_TASK 939 default 120 940 help 941 This option controls the default timeout (in seconds) used 942 to determine when a task has become non-responsive and should 943 be considered hung. 944 945 It can be adjusted at runtime via the kernel.hung_task_timeout_secs 946 sysctl or by writing a value to 947 /proc/sys/kernel/hung_task_timeout_secs. 948 949 A timeout of 0 disables the check. The default is two minutes. 950 Keeping the default should be fine in most cases. 951 952config BOOTPARAM_HUNG_TASK_PANIC 953 bool "Panic (Reboot) On Hung Tasks" 954 depends on DETECT_HUNG_TASK 955 help 956 Say Y here to enable the kernel to panic on "hung tasks", 957 which are bugs that cause the kernel to leave a task stuck 958 in uninterruptible "D" state. 959 960 The panic can be used in combination with panic_timeout, 961 to cause the system to reboot automatically after a 962 hung task has been detected. This feature is useful for 963 high-availability systems that have uptime guarantees and 964 where a hung tasks must be resolved ASAP. 965 966 Say N if unsure. 967 968config BOOTPARAM_HUNG_TASK_PANIC_VALUE 969 int 970 depends on DETECT_HUNG_TASK 971 range 0 1 972 default 0 if !BOOTPARAM_HUNG_TASK_PANIC 973 default 1 if BOOTPARAM_HUNG_TASK_PANIC 974 975config WQ_WATCHDOG 976 bool "Detect Workqueue Stalls" 977 depends on DEBUG_KERNEL 978 help 979 Say Y here to enable stall detection on workqueues. If a 980 worker pool doesn't make forward progress on a pending work 981 item for over a given amount of time, 30s by default, a 982 warning message is printed along with dump of workqueue 983 state. This can be configured through kernel parameter 984 "workqueue.watchdog_thresh" and its sysfs counterpart. 985 986config TEST_LOCKUP 987 tristate "Test module to generate lockups" 988 help 989 This builds the "test_lockup" module that helps to make sure 990 that watchdogs and lockup detectors are working properly. 991 992 Depending on module parameters it could emulate soft or hard 993 lockup, "hung task", or locking arbitrary lock for a long time. 994 Also it could generate series of lockups with cooling-down periods. 995 996 If unsure, say N. 997 998endmenu # "Debug lockups and hangs" 999 1000menu "Scheduler Debugging" 1001 1002config SCHED_DEBUG 1003 bool "Collect scheduler debugging info" 1004 depends on DEBUG_KERNEL && PROC_FS 1005 default y 1006 help 1007 If you say Y here, the /proc/sched_debug file will be provided 1008 that can help debug the scheduler. The runtime overhead of this 1009 option is minimal. 1010 1011config SCHED_INFO 1012 bool 1013 default n 1014 1015config SCHEDSTATS 1016 bool "Collect scheduler statistics" 1017 depends on DEBUG_KERNEL && PROC_FS 1018 select SCHED_INFO 1019 help 1020 If you say Y here, additional code will be inserted into the 1021 scheduler and related routines to collect statistics about 1022 scheduler behavior and provide them in /proc/schedstat. These 1023 stats may be useful for both tuning and debugging the scheduler 1024 If you aren't debugging the scheduler or trying to tune a specific 1025 application, you can say N to avoid the very slight overhead 1026 this adds. 1027 1028endmenu 1029 1030config DEBUG_TIMEKEEPING 1031 bool "Enable extra timekeeping sanity checking" 1032 help 1033 This option will enable additional timekeeping sanity checks 1034 which may be helpful when diagnosing issues where timekeeping 1035 problems are suspected. 1036 1037 This may include checks in the timekeeping hotpaths, so this 1038 option may have a (very small) performance impact to some 1039 workloads. 1040 1041 If unsure, say N. 1042 1043config DEBUG_PREEMPT 1044 bool "Debug preemptible kernel" 1045 depends on DEBUG_KERNEL && PREEMPTION && TRACE_IRQFLAGS_SUPPORT 1046 default y 1047 help 1048 If you say Y here then the kernel will use a debug variant of the 1049 commonly used smp_processor_id() function and will print warnings 1050 if kernel code uses it in a preemption-unsafe way. Also, the kernel 1051 will detect preemption count underflows. 1052 1053menu "Lock Debugging (spinlocks, mutexes, etc...)" 1054 1055config LOCK_DEBUGGING_SUPPORT 1056 bool 1057 depends on TRACE_IRQFLAGS_SUPPORT && STACKTRACE_SUPPORT && LOCKDEP_SUPPORT 1058 default y 1059 1060config PROVE_LOCKING 1061 bool "Lock debugging: prove locking correctness" 1062 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1063 select LOCKDEP 1064 select DEBUG_SPINLOCK 1065 select DEBUG_MUTEXES 1066 select DEBUG_RT_MUTEXES if RT_MUTEXES 1067 select DEBUG_RWSEMS 1068 select DEBUG_WW_MUTEX_SLOWPATH 1069 select DEBUG_LOCK_ALLOC 1070 select TRACE_IRQFLAGS 1071 default n 1072 help 1073 This feature enables the kernel to prove that all locking 1074 that occurs in the kernel runtime is mathematically 1075 correct: that under no circumstance could an arbitrary (and 1076 not yet triggered) combination of observed locking 1077 sequences (on an arbitrary number of CPUs, running an 1078 arbitrary number of tasks and interrupt contexts) cause a 1079 deadlock. 1080 1081 In short, this feature enables the kernel to report locking 1082 related deadlocks before they actually occur. 1083 1084 The proof does not depend on how hard and complex a 1085 deadlock scenario would be to trigger: how many 1086 participant CPUs, tasks and irq-contexts would be needed 1087 for it to trigger. The proof also does not depend on 1088 timing: if a race and a resulting deadlock is possible 1089 theoretically (no matter how unlikely the race scenario 1090 is), it will be proven so and will immediately be 1091 reported by the kernel (once the event is observed that 1092 makes the deadlock theoretically possible). 1093 1094 If a deadlock is impossible (i.e. the locking rules, as 1095 observed by the kernel, are mathematically correct), the 1096 kernel reports nothing. 1097 1098 NOTE: this feature can also be enabled for rwlocks, mutexes 1099 and rwsems - in which case all dependencies between these 1100 different locking variants are observed and mapped too, and 1101 the proof of observed correctness is also maintained for an 1102 arbitrary combination of these separate locking variants. 1103 1104 For more details, see Documentation/locking/lockdep-design.rst. 1105 1106config PROVE_RAW_LOCK_NESTING 1107 bool "Enable raw_spinlock - spinlock nesting checks" 1108 depends on PROVE_LOCKING 1109 default n 1110 help 1111 Enable the raw_spinlock vs. spinlock nesting checks which ensure 1112 that the lock nesting rules for PREEMPT_RT enabled kernels are 1113 not violated. 1114 1115 NOTE: There are known nesting problems. So if you enable this 1116 option expect lockdep splats until these problems have been fully 1117 addressed which is work in progress. This config switch allows to 1118 identify and analyze these problems. It will be removed and the 1119 check permanentely enabled once the main issues have been fixed. 1120 1121 If unsure, select N. 1122 1123config LOCK_STAT 1124 bool "Lock usage statistics" 1125 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1126 select LOCKDEP 1127 select DEBUG_SPINLOCK 1128 select DEBUG_MUTEXES 1129 select DEBUG_RT_MUTEXES if RT_MUTEXES 1130 select DEBUG_LOCK_ALLOC 1131 default n 1132 help 1133 This feature enables tracking lock contention points 1134 1135 For more details, see Documentation/locking/lockstat.rst 1136 1137 This also enables lock events required by "perf lock", 1138 subcommand of perf. 1139 If you want to use "perf lock", you also need to turn on 1140 CONFIG_EVENT_TRACING. 1141 1142 CONFIG_LOCK_STAT defines "contended" and "acquired" lock events. 1143 (CONFIG_LOCKDEP defines "acquire" and "release" events.) 1144 1145config DEBUG_RT_MUTEXES 1146 bool "RT Mutex debugging, deadlock detection" 1147 depends on DEBUG_KERNEL && RT_MUTEXES 1148 help 1149 This allows rt mutex semantics violations and rt mutex related 1150 deadlocks (lockups) to be detected and reported automatically. 1151 1152config DEBUG_SPINLOCK 1153 bool "Spinlock and rw-lock debugging: basic checks" 1154 depends on DEBUG_KERNEL 1155 select UNINLINE_SPIN_UNLOCK 1156 help 1157 Say Y here and build SMP to catch missing spinlock initialization 1158 and certain other kinds of spinlock errors commonly made. This is 1159 best used in conjunction with the NMI watchdog so that spinlock 1160 deadlocks are also debuggable. 1161 1162config DEBUG_MUTEXES 1163 bool "Mutex debugging: basic checks" 1164 depends on DEBUG_KERNEL 1165 help 1166 This feature allows mutex semantics violations to be detected and 1167 reported. 1168 1169config DEBUG_WW_MUTEX_SLOWPATH 1170 bool "Wait/wound mutex debugging: Slowpath testing" 1171 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1172 select DEBUG_LOCK_ALLOC 1173 select DEBUG_SPINLOCK 1174 select DEBUG_MUTEXES 1175 help 1176 This feature enables slowpath testing for w/w mutex users by 1177 injecting additional -EDEADLK wound/backoff cases. Together with 1178 the full mutex checks enabled with (CONFIG_PROVE_LOCKING) this 1179 will test all possible w/w mutex interface abuse with the 1180 exception of simply not acquiring all the required locks. 1181 Note that this feature can introduce significant overhead, so 1182 it really should not be enabled in a production or distro kernel, 1183 even a debug kernel. If you are a driver writer, enable it. If 1184 you are a distro, do not. 1185 1186config DEBUG_RWSEMS 1187 bool "RW Semaphore debugging: basic checks" 1188 depends on DEBUG_KERNEL 1189 help 1190 This debugging feature allows mismatched rw semaphore locks 1191 and unlocks to be detected and reported. 1192 1193config DEBUG_LOCK_ALLOC 1194 bool "Lock debugging: detect incorrect freeing of live locks" 1195 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1196 select DEBUG_SPINLOCK 1197 select DEBUG_MUTEXES 1198 select DEBUG_RT_MUTEXES if RT_MUTEXES 1199 select LOCKDEP 1200 help 1201 This feature will check whether any held lock (spinlock, rwlock, 1202 mutex or rwsem) is incorrectly freed by the kernel, via any of the 1203 memory-freeing routines (kfree(), kmem_cache_free(), free_pages(), 1204 vfree(), etc.), whether a live lock is incorrectly reinitialized via 1205 spin_lock_init()/mutex_init()/etc., or whether there is any lock 1206 held during task exit. 1207 1208config LOCKDEP 1209 bool 1210 depends on DEBUG_KERNEL && LOCK_DEBUGGING_SUPPORT 1211 select STACKTRACE 1212 select FRAME_POINTER if !MIPS && !PPC && !ARM && !S390 && !MICROBLAZE && !ARC && !X86 1213 select KALLSYMS 1214 select KALLSYMS_ALL 1215 1216config LOCKDEP_SMALL 1217 bool 1218 1219config DEBUG_LOCKDEP 1220 bool "Lock dependency engine debugging" 1221 depends on DEBUG_KERNEL && LOCKDEP 1222 help 1223 If you say Y here, the lock dependency engine will do 1224 additional runtime checks to debug itself, at the price 1225 of more runtime overhead. 1226 1227config DEBUG_ATOMIC_SLEEP 1228 bool "Sleep inside atomic section checking" 1229 select PREEMPT_COUNT 1230 depends on DEBUG_KERNEL 1231 depends on !ARCH_NO_PREEMPT 1232 help 1233 If you say Y here, various routines which may sleep will become very 1234 noisy if they are called inside atomic sections: when a spinlock is 1235 held, inside an rcu read side critical section, inside preempt disabled 1236 sections, inside an interrupt, etc... 1237 1238config DEBUG_LOCKING_API_SELFTESTS 1239 bool "Locking API boot-time self-tests" 1240 depends on DEBUG_KERNEL 1241 help 1242 Say Y here if you want the kernel to run a short self-test during 1243 bootup. The self-test checks whether common types of locking bugs 1244 are detected by debugging mechanisms or not. (if you disable 1245 lock debugging then those bugs wont be detected of course.) 1246 The following locking APIs are covered: spinlocks, rwlocks, 1247 mutexes and rwsems. 1248 1249config LOCK_TORTURE_TEST 1250 tristate "torture tests for locking" 1251 depends on DEBUG_KERNEL 1252 select TORTURE_TEST 1253 help 1254 This option provides a kernel module that runs torture tests 1255 on kernel locking primitives. The kernel module may be built 1256 after the fact on the running kernel to be tested, if desired. 1257 1258 Say Y here if you want kernel locking-primitive torture tests 1259 to be built into the kernel. 1260 Say M if you want these torture tests to build as a module. 1261 Say N if you are unsure. 1262 1263config WW_MUTEX_SELFTEST 1264 tristate "Wait/wound mutex selftests" 1265 help 1266 This option provides a kernel module that runs tests on the 1267 on the struct ww_mutex locking API. 1268 1269 It is recommended to enable DEBUG_WW_MUTEX_SLOWPATH in conjunction 1270 with this test harness. 1271 1272 Say M if you want these self tests to build as a module. 1273 Say N if you are unsure. 1274 1275endmenu # lock debugging 1276 1277config TRACE_IRQFLAGS 1278 bool 1279 help 1280 Enables hooks to interrupt enabling and disabling for 1281 either tracing or lock debugging. 1282 1283config STACKTRACE 1284 bool "Stack backtrace support" 1285 depends on STACKTRACE_SUPPORT 1286 help 1287 This option causes the kernel to create a /proc/pid/stack for 1288 every process, showing its current stack trace. 1289 It is also used by various kernel debugging features that require 1290 stack trace generation. 1291 1292config WARN_ALL_UNSEEDED_RANDOM 1293 bool "Warn for all uses of unseeded randomness" 1294 default n 1295 help 1296 Some parts of the kernel contain bugs relating to their use of 1297 cryptographically secure random numbers before it's actually possible 1298 to generate those numbers securely. This setting ensures that these 1299 flaws don't go unnoticed, by enabling a message, should this ever 1300 occur. This will allow people with obscure setups to know when things 1301 are going wrong, so that they might contact developers about fixing 1302 it. 1303 1304 Unfortunately, on some models of some architectures getting 1305 a fully seeded CRNG is extremely difficult, and so this can 1306 result in dmesg getting spammed for a surprisingly long 1307 time. This is really bad from a security perspective, and 1308 so architecture maintainers really need to do what they can 1309 to get the CRNG seeded sooner after the system is booted. 1310 However, since users cannot do anything actionable to 1311 address this, by default the kernel will issue only a single 1312 warning for the first use of unseeded randomness. 1313 1314 Say Y here if you want to receive warnings for all uses of 1315 unseeded randomness. This will be of use primarily for 1316 those developers interested in improving the security of 1317 Linux kernels running on their architecture (or 1318 subarchitecture). 1319 1320config DEBUG_KOBJECT 1321 bool "kobject debugging" 1322 depends on DEBUG_KERNEL 1323 help 1324 If you say Y here, some extra kobject debugging messages will be sent 1325 to the syslog. 1326 1327config DEBUG_KOBJECT_RELEASE 1328 bool "kobject release debugging" 1329 depends on DEBUG_OBJECTS_TIMERS 1330 help 1331 kobjects are reference counted objects. This means that their 1332 last reference count put is not predictable, and the kobject can 1333 live on past the point at which a driver decides to drop it's 1334 initial reference to the kobject gained on allocation. An 1335 example of this would be a struct device which has just been 1336 unregistered. 1337 1338 However, some buggy drivers assume that after such an operation, 1339 the memory backing the kobject can be immediately freed. This 1340 goes completely against the principles of a refcounted object. 1341 1342 If you say Y here, the kernel will delay the release of kobjects 1343 on the last reference count to improve the visibility of this 1344 kind of kobject release bug. 1345 1346config HAVE_DEBUG_BUGVERBOSE 1347 bool 1348 1349menu "Debug kernel data structures" 1350 1351config DEBUG_LIST 1352 bool "Debug linked list manipulation" 1353 depends on DEBUG_KERNEL || BUG_ON_DATA_CORRUPTION 1354 help 1355 Enable this to turn on extended checks in the linked-list 1356 walking routines. 1357 1358 If unsure, say N. 1359 1360config DEBUG_PLIST 1361 bool "Debug priority linked list manipulation" 1362 depends on DEBUG_KERNEL 1363 help 1364 Enable this to turn on extended checks in the priority-ordered 1365 linked-list (plist) walking routines. This checks the entire 1366 list multiple times during each manipulation. 1367 1368 If unsure, say N. 1369 1370config DEBUG_SG 1371 bool "Debug SG table operations" 1372 depends on DEBUG_KERNEL 1373 help 1374 Enable this to turn on checks on scatter-gather tables. This can 1375 help find problems with drivers that do not properly initialize 1376 their sg tables. 1377 1378 If unsure, say N. 1379 1380config DEBUG_NOTIFIERS 1381 bool "Debug notifier call chains" 1382 depends on DEBUG_KERNEL 1383 help 1384 Enable this to turn on sanity checking for notifier call chains. 1385 This is most useful for kernel developers to make sure that 1386 modules properly unregister themselves from notifier chains. 1387 This is a relatively cheap check but if you care about maximum 1388 performance, say N. 1389 1390config BUG_ON_DATA_CORRUPTION 1391 bool "Trigger a BUG when data corruption is detected" 1392 select DEBUG_LIST 1393 help 1394 Select this option if the kernel should BUG when it encounters 1395 data corruption in kernel memory structures when they get checked 1396 for validity. 1397 1398 If unsure, say N. 1399 1400endmenu 1401 1402config DEBUG_CREDENTIALS 1403 bool "Debug credential management" 1404 depends on DEBUG_KERNEL 1405 help 1406 Enable this to turn on some debug checking for credential 1407 management. The additional code keeps track of the number of 1408 pointers from task_structs to any given cred struct, and checks to 1409 see that this number never exceeds the usage count of the cred 1410 struct. 1411 1412 Furthermore, if SELinux is enabled, this also checks that the 1413 security pointer in the cred struct is never seen to be invalid. 1414 1415 If unsure, say N. 1416 1417source "kernel/rcu/Kconfig.debug" 1418 1419config DEBUG_WQ_FORCE_RR_CPU 1420 bool "Force round-robin CPU selection for unbound work items" 1421 depends on DEBUG_KERNEL 1422 default n 1423 help 1424 Workqueue used to implicitly guarantee that work items queued 1425 without explicit CPU specified are put on the local CPU. This 1426 guarantee is no longer true and while local CPU is still 1427 preferred work items may be put on foreign CPUs. Kernel 1428 parameter "workqueue.debug_force_rr_cpu" is added to force 1429 round-robin CPU selection to flush out usages which depend on the 1430 now broken guarantee. This config option enables the debug 1431 feature by default. When enabled, memory and cache locality will 1432 be impacted. 1433 1434config DEBUG_BLOCK_EXT_DEVT 1435 bool "Force extended block device numbers and spread them" 1436 depends on DEBUG_KERNEL 1437 depends on BLOCK 1438 default n 1439 help 1440 BIG FAT WARNING: ENABLING THIS OPTION MIGHT BREAK BOOTING ON 1441 SOME DISTRIBUTIONS. DO NOT ENABLE THIS UNLESS YOU KNOW WHAT 1442 YOU ARE DOING. Distros, please enable this and fix whatever 1443 is broken. 1444 1445 Conventionally, block device numbers are allocated from 1446 predetermined contiguous area. However, extended block area 1447 may introduce non-contiguous block device numbers. This 1448 option forces most block device numbers to be allocated from 1449 the extended space and spreads them to discover kernel or 1450 userland code paths which assume predetermined contiguous 1451 device number allocation. 1452 1453 Note that turning on this debug option shuffles all the 1454 device numbers for all IDE and SCSI devices including libata 1455 ones, so root partition specified using device number 1456 directly (via rdev or root=MAJ:MIN) won't work anymore. 1457 Textual device names (root=/dev/sdXn) will continue to work. 1458 1459 Say N if you are unsure. 1460 1461config CPU_HOTPLUG_STATE_CONTROL 1462 bool "Enable CPU hotplug state control" 1463 depends on DEBUG_KERNEL 1464 depends on HOTPLUG_CPU 1465 default n 1466 help 1467 Allows to write steps between "offline" and "online" to the CPUs 1468 sysfs target file so states can be stepped granular. This is a debug 1469 option for now as the hotplug machinery cannot be stopped and 1470 restarted at arbitrary points yet. 1471 1472 Say N if your are unsure. 1473 1474config LATENCYTOP 1475 bool "Latency measuring infrastructure" 1476 depends on DEBUG_KERNEL 1477 depends on STACKTRACE_SUPPORT 1478 depends on PROC_FS 1479 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86 1480 select KALLSYMS 1481 select KALLSYMS_ALL 1482 select STACKTRACE 1483 select SCHEDSTATS 1484 select SCHED_DEBUG 1485 help 1486 Enable this option if you want to use the LatencyTOP tool 1487 to find out which userspace is blocking on what kernel operations. 1488 1489source "kernel/trace/Kconfig" 1490 1491config PROVIDE_OHCI1394_DMA_INIT 1492 bool "Remote debugging over FireWire early on boot" 1493 depends on PCI && X86 1494 help 1495 If you want to debug problems which hang or crash the kernel early 1496 on boot and the crashing machine has a FireWire port, you can use 1497 this feature to remotely access the memory of the crashed machine 1498 over FireWire. This employs remote DMA as part of the OHCI1394 1499 specification which is now the standard for FireWire controllers. 1500 1501 With remote DMA, you can monitor the printk buffer remotely using 1502 firescope and access all memory below 4GB using fireproxy from gdb. 1503 Even controlling a kernel debugger is possible using remote DMA. 1504 1505 Usage: 1506 1507 If ohci1394_dma=early is used as boot parameter, it will initialize 1508 all OHCI1394 controllers which are found in the PCI config space. 1509 1510 As all changes to the FireWire bus such as enabling and disabling 1511 devices cause a bus reset and thereby disable remote DMA for all 1512 devices, be sure to have the cable plugged and FireWire enabled on 1513 the debugging host before booting the debug target for debugging. 1514 1515 This code (~1k) is freed after boot. By then, the firewire stack 1516 in charge of the OHCI-1394 controllers should be used instead. 1517 1518 See Documentation/core-api/debugging-via-ohci1394.rst for more information. 1519 1520source "samples/Kconfig" 1521 1522config ARCH_HAS_DEVMEM_IS_ALLOWED 1523 bool 1524 1525config STRICT_DEVMEM 1526 bool "Filter access to /dev/mem" 1527 depends on MMU && DEVMEM 1528 depends on ARCH_HAS_DEVMEM_IS_ALLOWED 1529 default y if PPC || X86 || ARM64 1530 help 1531 If this option is disabled, you allow userspace (root) access to all 1532 of memory, including kernel and userspace memory. Accidental 1533 access to this is obviously disastrous, but specific access can 1534 be used by people debugging the kernel. Note that with PAT support 1535 enabled, even in this case there are restrictions on /dev/mem 1536 use due to the cache aliasing requirements. 1537 1538 If this option is switched on, and IO_STRICT_DEVMEM=n, the /dev/mem 1539 file only allows userspace access to PCI space and the BIOS code and 1540 data regions. This is sufficient for dosemu and X and all common 1541 users of /dev/mem. 1542 1543 If in doubt, say Y. 1544 1545config IO_STRICT_DEVMEM 1546 bool "Filter I/O access to /dev/mem" 1547 depends on STRICT_DEVMEM 1548 help 1549 If this option is disabled, you allow userspace (root) access to all 1550 io-memory regardless of whether a driver is actively using that 1551 range. Accidental access to this is obviously disastrous, but 1552 specific access can be used by people debugging kernel drivers. 1553 1554 If this option is switched on, the /dev/mem file only allows 1555 userspace access to *idle* io-memory ranges (see /proc/iomem) This 1556 may break traditional users of /dev/mem (dosemu, legacy X, etc...) 1557 if the driver using a given range cannot be disabled. 1558 1559 If in doubt, say Y. 1560 1561menu "$(SRCARCH) Debugging" 1562 1563source "arch/$(SRCARCH)/Kconfig.debug" 1564 1565endmenu 1566 1567menu "Kernel Testing and Coverage" 1568 1569source "lib/kunit/Kconfig" 1570 1571config NOTIFIER_ERROR_INJECTION 1572 tristate "Notifier error injection" 1573 depends on DEBUG_KERNEL 1574 select DEBUG_FS 1575 help 1576 This option provides the ability to inject artificial errors to 1577 specified notifier chain callbacks. It is useful to test the error 1578 handling of notifier call chain failures. 1579 1580 Say N if unsure. 1581 1582config PM_NOTIFIER_ERROR_INJECT 1583 tristate "PM notifier error injection module" 1584 depends on PM && NOTIFIER_ERROR_INJECTION 1585 default m if PM_DEBUG 1586 help 1587 This option provides the ability to inject artificial errors to 1588 PM notifier chain callbacks. It is controlled through debugfs 1589 interface /sys/kernel/debug/notifier-error-inject/pm 1590 1591 If the notifier call chain should be failed with some events 1592 notified, write the error code to "actions/<notifier event>/error". 1593 1594 Example: Inject PM suspend error (-12 = -ENOMEM) 1595 1596 # cd /sys/kernel/debug/notifier-error-inject/pm/ 1597 # echo -12 > actions/PM_SUSPEND_PREPARE/error 1598 # echo mem > /sys/power/state 1599 bash: echo: write error: Cannot allocate memory 1600 1601 To compile this code as a module, choose M here: the module will 1602 be called pm-notifier-error-inject. 1603 1604 If unsure, say N. 1605 1606config OF_RECONFIG_NOTIFIER_ERROR_INJECT 1607 tristate "OF reconfig notifier error injection module" 1608 depends on OF_DYNAMIC && NOTIFIER_ERROR_INJECTION 1609 help 1610 This option provides the ability to inject artificial errors to 1611 OF reconfig notifier chain callbacks. It is controlled 1612 through debugfs interface under 1613 /sys/kernel/debug/notifier-error-inject/OF-reconfig/ 1614 1615 If the notifier call chain should be failed with some events 1616 notified, write the error code to "actions/<notifier event>/error". 1617 1618 To compile this code as a module, choose M here: the module will 1619 be called of-reconfig-notifier-error-inject. 1620 1621 If unsure, say N. 1622 1623config NETDEV_NOTIFIER_ERROR_INJECT 1624 tristate "Netdev notifier error injection module" 1625 depends on NET && NOTIFIER_ERROR_INJECTION 1626 help 1627 This option provides the ability to inject artificial errors to 1628 netdevice notifier chain callbacks. It is controlled through debugfs 1629 interface /sys/kernel/debug/notifier-error-inject/netdev 1630 1631 If the notifier call chain should be failed with some events 1632 notified, write the error code to "actions/<notifier event>/error". 1633 1634 Example: Inject netdevice mtu change error (-22 = -EINVAL) 1635 1636 # cd /sys/kernel/debug/notifier-error-inject/netdev 1637 # echo -22 > actions/NETDEV_CHANGEMTU/error 1638 # ip link set eth0 mtu 1024 1639 RTNETLINK answers: Invalid argument 1640 1641 To compile this code as a module, choose M here: the module will 1642 be called netdev-notifier-error-inject. 1643 1644 If unsure, say N. 1645 1646config FUNCTION_ERROR_INJECTION 1647 def_bool y 1648 depends on HAVE_FUNCTION_ERROR_INJECTION && KPROBES 1649 1650config FAULT_INJECTION 1651 bool "Fault-injection framework" 1652 depends on DEBUG_KERNEL 1653 help 1654 Provide fault-injection framework. 1655 For more details, see Documentation/fault-injection/. 1656 1657config FAILSLAB 1658 bool "Fault-injection capability for kmalloc" 1659 depends on FAULT_INJECTION 1660 depends on SLAB || SLUB 1661 help 1662 Provide fault-injection capability for kmalloc. 1663 1664config FAIL_PAGE_ALLOC 1665 bool "Fault-injection capability for alloc_pages()" 1666 depends on FAULT_INJECTION 1667 help 1668 Provide fault-injection capability for alloc_pages(). 1669 1670config FAIL_MAKE_REQUEST 1671 bool "Fault-injection capability for disk IO" 1672 depends on FAULT_INJECTION && BLOCK 1673 help 1674 Provide fault-injection capability for disk IO. 1675 1676config FAIL_IO_TIMEOUT 1677 bool "Fault-injection capability for faking disk interrupts" 1678 depends on FAULT_INJECTION && BLOCK 1679 help 1680 Provide fault-injection capability on end IO handling. This 1681 will make the block layer "forget" an interrupt as configured, 1682 thus exercising the error handling. 1683 1684 Only works with drivers that use the generic timeout handling, 1685 for others it wont do anything. 1686 1687config FAIL_FUTEX 1688 bool "Fault-injection capability for futexes" 1689 select DEBUG_FS 1690 depends on FAULT_INJECTION && FUTEX 1691 help 1692 Provide fault-injection capability for futexes. 1693 1694config FAULT_INJECTION_DEBUG_FS 1695 bool "Debugfs entries for fault-injection capabilities" 1696 depends on FAULT_INJECTION && SYSFS && DEBUG_FS 1697 help 1698 Enable configuration of fault-injection capabilities via debugfs. 1699 1700config FAIL_FUNCTION 1701 bool "Fault-injection capability for functions" 1702 depends on FAULT_INJECTION_DEBUG_FS && FUNCTION_ERROR_INJECTION 1703 help 1704 Provide function-based fault-injection capability. 1705 This will allow you to override a specific function with a return 1706 with given return value. As a result, function caller will see 1707 an error value and have to handle it. This is useful to test the 1708 error handling in various subsystems. 1709 1710config FAIL_MMC_REQUEST 1711 bool "Fault-injection capability for MMC IO" 1712 depends on FAULT_INJECTION_DEBUG_FS && MMC 1713 help 1714 Provide fault-injection capability for MMC IO. 1715 This will make the mmc core return data errors. This is 1716 useful to test the error handling in the mmc block device 1717 and to test how the mmc host driver handles retries from 1718 the block device. 1719 1720config FAULT_INJECTION_STACKTRACE_FILTER 1721 bool "stacktrace filter for fault-injection capabilities" 1722 depends on FAULT_INJECTION_DEBUG_FS && STACKTRACE_SUPPORT 1723 depends on !X86_64 1724 select STACKTRACE 1725 select FRAME_POINTER if !MIPS && !PPC && !S390 && !MICROBLAZE && !ARM && !ARC && !X86 1726 help 1727 Provide stacktrace filter for fault-injection capabilities 1728 1729config ARCH_HAS_KCOV 1730 bool 1731 help 1732 An architecture should select this when it can successfully 1733 build and run with CONFIG_KCOV. This typically requires 1734 disabling instrumentation for some early boot code. 1735 1736config CC_HAS_SANCOV_TRACE_PC 1737 def_bool $(cc-option,-fsanitize-coverage=trace-pc) 1738 1739 1740config KCOV 1741 bool "Code coverage for fuzzing" 1742 depends on ARCH_HAS_KCOV 1743 depends on CC_HAS_SANCOV_TRACE_PC || GCC_PLUGINS 1744 select DEBUG_FS 1745 select GCC_PLUGIN_SANCOV if !CC_HAS_SANCOV_TRACE_PC 1746 help 1747 KCOV exposes kernel code coverage information in a form suitable 1748 for coverage-guided fuzzing (randomized testing). 1749 1750 If RANDOMIZE_BASE is enabled, PC values will not be stable across 1751 different machines and across reboots. If you need stable PC values, 1752 disable RANDOMIZE_BASE. 1753 1754 For more details, see Documentation/dev-tools/kcov.rst. 1755 1756config KCOV_ENABLE_COMPARISONS 1757 bool "Enable comparison operands collection by KCOV" 1758 depends on KCOV 1759 depends on $(cc-option,-fsanitize-coverage=trace-cmp) 1760 help 1761 KCOV also exposes operands of every comparison in the instrumented 1762 code along with operand sizes and PCs of the comparison instructions. 1763 These operands can be used by fuzzing engines to improve the quality 1764 of fuzzing coverage. 1765 1766config KCOV_INSTRUMENT_ALL 1767 bool "Instrument all code by default" 1768 depends on KCOV 1769 default y 1770 help 1771 If you are doing generic system call fuzzing (like e.g. syzkaller), 1772 then you will want to instrument the whole kernel and you should 1773 say y here. If you are doing more targeted fuzzing (like e.g. 1774 filesystem fuzzing with AFL) then you will want to enable coverage 1775 for more specific subsets of files, and should say n here. 1776 1777menuconfig RUNTIME_TESTING_MENU 1778 bool "Runtime Testing" 1779 def_bool y 1780 1781if RUNTIME_TESTING_MENU 1782 1783config LKDTM 1784 tristate "Linux Kernel Dump Test Tool Module" 1785 depends on DEBUG_FS 1786 help 1787 This module enables testing of the different dumping mechanisms by 1788 inducing system failures at predefined crash points. 1789 If you don't need it: say N 1790 Choose M here to compile this code as a module. The module will be 1791 called lkdtm. 1792 1793 Documentation on how to use the module can be found in 1794 Documentation/fault-injection/provoke-crashes.rst 1795 1796config TEST_LIST_SORT 1797 tristate "Linked list sorting test" 1798 depends on DEBUG_KERNEL || m 1799 help 1800 Enable this to turn on 'list_sort()' function test. This test is 1801 executed only once during system boot (so affects only boot time), 1802 or at module load time. 1803 1804 If unsure, say N. 1805 1806config TEST_MIN_HEAP 1807 tristate "Min heap test" 1808 depends on DEBUG_KERNEL || m 1809 help 1810 Enable this to turn on min heap function tests. This test is 1811 executed only once during system boot (so affects only boot time), 1812 or at module load time. 1813 1814 If unsure, say N. 1815 1816config TEST_SORT 1817 tristate "Array-based sort test" 1818 depends on DEBUG_KERNEL || m 1819 help 1820 This option enables the self-test function of 'sort()' at boot, 1821 or at module load time. 1822 1823 If unsure, say N. 1824 1825config KPROBES_SANITY_TEST 1826 bool "Kprobes sanity tests" 1827 depends on DEBUG_KERNEL 1828 depends on KPROBES 1829 help 1830 This option provides for testing basic kprobes functionality on 1831 boot. Samples of kprobe and kretprobe are inserted and 1832 verified for functionality. 1833 1834 Say N if you are unsure. 1835 1836config BACKTRACE_SELF_TEST 1837 tristate "Self test for the backtrace code" 1838 depends on DEBUG_KERNEL 1839 help 1840 This option provides a kernel module that can be used to test 1841 the kernel stack backtrace code. This option is not useful 1842 for distributions or general kernels, but only for kernel 1843 developers working on architecture code. 1844 1845 Note that if you want to also test saved backtraces, you will 1846 have to enable STACKTRACE as well. 1847 1848 Say N if you are unsure. 1849 1850config RBTREE_TEST 1851 tristate "Red-Black tree test" 1852 depends on DEBUG_KERNEL 1853 help 1854 A benchmark measuring the performance of the rbtree library. 1855 Also includes rbtree invariant checks. 1856 1857config REED_SOLOMON_TEST 1858 tristate "Reed-Solomon library test" 1859 depends on DEBUG_KERNEL || m 1860 select REED_SOLOMON 1861 select REED_SOLOMON_ENC16 1862 select REED_SOLOMON_DEC16 1863 help 1864 This option enables the self-test function of rslib at boot, 1865 or at module load time. 1866 1867 If unsure, say N. 1868 1869config INTERVAL_TREE_TEST 1870 tristate "Interval tree test" 1871 depends on DEBUG_KERNEL 1872 select INTERVAL_TREE 1873 help 1874 A benchmark measuring the performance of the interval tree library 1875 1876config PERCPU_TEST 1877 tristate "Per cpu operations test" 1878 depends on m && DEBUG_KERNEL 1879 help 1880 Enable this option to build test module which validates per-cpu 1881 operations. 1882 1883 If unsure, say N. 1884 1885config ATOMIC64_SELFTEST 1886 tristate "Perform an atomic64_t self-test" 1887 help 1888 Enable this option to test the atomic64_t functions at boot or 1889 at module load time. 1890 1891 If unsure, say N. 1892 1893config ASYNC_RAID6_TEST 1894 tristate "Self test for hardware accelerated raid6 recovery" 1895 depends on ASYNC_RAID6_RECOV 1896 select ASYNC_MEMCPY 1897 ---help--- 1898 This is a one-shot self test that permutes through the 1899 recovery of all the possible two disk failure scenarios for a 1900 N-disk array. Recovery is performed with the asynchronous 1901 raid6 recovery routines, and will optionally use an offload 1902 engine if one is available. 1903 1904 If unsure, say N. 1905 1906config TEST_HEXDUMP 1907 tristate "Test functions located in the hexdump module at runtime" 1908 1909config TEST_STRING_HELPERS 1910 tristate "Test functions located in the string_helpers module at runtime" 1911 1912config TEST_STRSCPY 1913 tristate "Test strscpy*() family of functions at runtime" 1914 1915config TEST_KSTRTOX 1916 tristate "Test kstrto*() family of functions at runtime" 1917 1918config TEST_PRINTF 1919 tristate "Test printf() family of functions at runtime" 1920 1921config TEST_BITMAP 1922 tristate "Test bitmap_*() family of functions at runtime" 1923 help 1924 Enable this option to test the bitmap functions at boot. 1925 1926 If unsure, say N. 1927 1928config TEST_BITFIELD 1929 tristate "Test bitfield functions at runtime" 1930 help 1931 Enable this option to test the bitfield functions at boot. 1932 1933 If unsure, say N. 1934 1935config TEST_UUID 1936 tristate "Test functions located in the uuid module at runtime" 1937 1938config TEST_XARRAY 1939 tristate "Test the XArray code at runtime" 1940 1941config TEST_OVERFLOW 1942 tristate "Test check_*_overflow() functions at runtime" 1943 1944config TEST_RHASHTABLE 1945 tristate "Perform selftest on resizable hash table" 1946 help 1947 Enable this option to test the rhashtable functions at boot. 1948 1949 If unsure, say N. 1950 1951config TEST_HASH 1952 tristate "Perform selftest on hash functions" 1953 help 1954 Enable this option to test the kernel's integer (<linux/hash.h>), 1955 string (<linux/stringhash.h>), and siphash (<linux/siphash.h>) 1956 hash functions on boot (or module load). 1957 1958 This is intended to help people writing architecture-specific 1959 optimized versions. If unsure, say N. 1960 1961config TEST_IDA 1962 tristate "Perform selftest on IDA functions" 1963 1964config TEST_PARMAN 1965 tristate "Perform selftest on priority array manager" 1966 depends on PARMAN 1967 help 1968 Enable this option to test priority array manager on boot 1969 (or module load). 1970 1971 If unsure, say N. 1972 1973config TEST_IRQ_TIMINGS 1974 bool "IRQ timings selftest" 1975 depends on IRQ_TIMINGS 1976 help 1977 Enable this option to test the irq timings code on boot. 1978 1979 If unsure, say N. 1980 1981config TEST_LKM 1982 tristate "Test module loading with 'hello world' module" 1983 depends on m 1984 help 1985 This builds the "test_module" module that emits "Hello, world" 1986 on printk when loaded. It is designed to be used for basic 1987 evaluation of the module loading subsystem (for example when 1988 validating module verification). It lacks any extra dependencies, 1989 and will not normally be loaded by the system unless explicitly 1990 requested by name. 1991 1992 If unsure, say N. 1993 1994config TEST_VMALLOC 1995 tristate "Test module for stress/performance analysis of vmalloc allocator" 1996 default n 1997 depends on MMU 1998 depends on m 1999 help 2000 This builds the "test_vmalloc" module that should be used for 2001 stress and performance analysis. So, any new change for vmalloc 2002 subsystem can be evaluated from performance and stability point 2003 of view. 2004 2005 If unsure, say N. 2006 2007config TEST_USER_COPY 2008 tristate "Test user/kernel boundary protections" 2009 depends on m 2010 help 2011 This builds the "test_user_copy" module that runs sanity checks 2012 on the copy_to/from_user infrastructure, making sure basic 2013 user/kernel boundary testing is working. If it fails to load, 2014 a regression has been detected in the user/kernel memory boundary 2015 protections. 2016 2017 If unsure, say N. 2018 2019config TEST_BPF 2020 tristate "Test BPF filter functionality" 2021 depends on m && NET 2022 help 2023 This builds the "test_bpf" module that runs various test vectors 2024 against the BPF interpreter or BPF JIT compiler depending on the 2025 current setting. This is in particular useful for BPF JIT compiler 2026 development, but also to run regression tests against changes in 2027 the interpreter code. It also enables test stubs for eBPF maps and 2028 verifier used by user space verifier testsuite. 2029 2030 If unsure, say N. 2031 2032config TEST_BLACKHOLE_DEV 2033 tristate "Test blackhole netdev functionality" 2034 depends on m && NET 2035 help 2036 This builds the "test_blackhole_dev" module that validates the 2037 data path through this blackhole netdev. 2038 2039 If unsure, say N. 2040 2041config FIND_BIT_BENCHMARK 2042 tristate "Test find_bit functions" 2043 help 2044 This builds the "test_find_bit" module that measure find_*_bit() 2045 functions performance. 2046 2047 If unsure, say N. 2048 2049config TEST_FIRMWARE 2050 tristate "Test firmware loading via userspace interface" 2051 depends on FW_LOADER 2052 help 2053 This builds the "test_firmware" module that creates a userspace 2054 interface for testing firmware loading. This can be used to 2055 control the triggering of firmware loading without needing an 2056 actual firmware-using device. The contents can be rechecked by 2057 userspace. 2058 2059 If unsure, say N. 2060 2061config TEST_SYSCTL 2062 tristate "sysctl test driver" 2063 depends on PROC_SYSCTL 2064 help 2065 This builds the "test_sysctl" module. This driver enables to test the 2066 proc sysctl interfaces available to drivers safely without affecting 2067 production knobs which might alter system functionality. 2068 2069 If unsure, say N. 2070 2071config SYSCTL_KUNIT_TEST 2072 tristate "KUnit test for sysctl" 2073 depends on KUNIT 2074 help 2075 This builds the proc sysctl unit test, which runs on boot. 2076 Tests the API contract and implementation correctness of sysctl. 2077 For more information on KUnit and unit tests in general please refer 2078 to the KUnit documentation in Documentation/dev-tools/kunit/. 2079 2080 If unsure, say N. 2081 2082config LIST_KUNIT_TEST 2083 tristate "KUnit Test for Kernel Linked-list structures" 2084 depends on KUNIT 2085 help 2086 This builds the linked list KUnit test suite. 2087 It tests that the API and basic functionality of the list_head type 2088 and associated macros. 2089 2090 KUnit tests run during boot and output the results to the debug log 2091 in TAP format (http://testanything.org/). Only useful for kernel devs 2092 running the KUnit test harness, and not intended for inclusion into a 2093 production build. 2094 2095 For more information on KUnit and unit tests in general please refer 2096 to the KUnit documentation in Documentation/dev-tools/kunit/. 2097 2098 If unsure, say N. 2099 2100config LINEAR_RANGES_TEST 2101 tristate "KUnit test for linear_ranges" 2102 depends on KUNIT 2103 select LINEAR_RANGES 2104 help 2105 This builds the linear_ranges unit test, which runs on boot. 2106 Tests the linear_ranges logic correctness. 2107 For more information on KUnit and unit tests in general please refer 2108 to the KUnit documentation in Documentation/dev-tools/kunit/. 2109 2110 If unsure, say N. 2111 2112config TEST_UDELAY 2113 tristate "udelay test driver" 2114 help 2115 This builds the "udelay_test" module that helps to make sure 2116 that udelay() is working properly. 2117 2118 If unsure, say N. 2119 2120config TEST_STATIC_KEYS 2121 tristate "Test static keys" 2122 depends on m 2123 help 2124 Test the static key interfaces. 2125 2126 If unsure, say N. 2127 2128config TEST_KMOD 2129 tristate "kmod stress tester" 2130 depends on m 2131 depends on NETDEVICES && NET_CORE && INET # for TUN 2132 depends on BLOCK 2133 select TEST_LKM 2134 select XFS_FS 2135 select TUN 2136 select BTRFS_FS 2137 help 2138 Test the kernel's module loading mechanism: kmod. kmod implements 2139 support to load modules using the Linux kernel's usermode helper. 2140 This test provides a series of tests against kmod. 2141 2142 Although technically you can either build test_kmod as a module or 2143 into the kernel we disallow building it into the kernel since 2144 it stress tests request_module() and this will very likely cause 2145 some issues by taking over precious threads available from other 2146 module load requests, ultimately this could be fatal. 2147 2148 To run tests run: 2149 2150 tools/testing/selftests/kmod/kmod.sh --help 2151 2152 If unsure, say N. 2153 2154config TEST_DEBUG_VIRTUAL 2155 tristate "Test CONFIG_DEBUG_VIRTUAL feature" 2156 depends on DEBUG_VIRTUAL 2157 help 2158 Test the kernel's ability to detect incorrect calls to 2159 virt_to_phys() done against the non-linear part of the 2160 kernel's virtual address map. 2161 2162 If unsure, say N. 2163 2164config TEST_MEMCAT_P 2165 tristate "Test memcat_p() helper function" 2166 help 2167 Test the memcat_p() helper for correctly merging two 2168 pointer arrays together. 2169 2170 If unsure, say N. 2171 2172config TEST_LIVEPATCH 2173 tristate "Test livepatching" 2174 default n 2175 depends on DYNAMIC_DEBUG 2176 depends on LIVEPATCH 2177 depends on m 2178 help 2179 Test kernel livepatching features for correctness. The tests will 2180 load test modules that will be livepatched in various scenarios. 2181 2182 To run all the livepatching tests: 2183 2184 make -C tools/testing/selftests TARGETS=livepatch run_tests 2185 2186 Alternatively, individual tests may be invoked: 2187 2188 tools/testing/selftests/livepatch/test-callbacks.sh 2189 tools/testing/selftests/livepatch/test-livepatch.sh 2190 tools/testing/selftests/livepatch/test-shadow-vars.sh 2191 2192 If unsure, say N. 2193 2194config TEST_OBJAGG 2195 tristate "Perform selftest on object aggreration manager" 2196 default n 2197 depends on OBJAGG 2198 help 2199 Enable this option to test object aggregation manager on boot 2200 (or module load). 2201 2202 2203config TEST_STACKINIT 2204 tristate "Test level of stack variable initialization" 2205 help 2206 Test if the kernel is zero-initializing stack variables and 2207 padding. Coverage is controlled by compiler flags, 2208 CONFIG_GCC_PLUGIN_STRUCTLEAK, CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF, 2209 or CONFIG_GCC_PLUGIN_STRUCTLEAK_BYREF_ALL. 2210 2211 If unsure, say N. 2212 2213config TEST_MEMINIT 2214 tristate "Test heap/page initialization" 2215 help 2216 Test if the kernel is zero-initializing heap and page allocations. 2217 This can be useful to test init_on_alloc and init_on_free features. 2218 2219 If unsure, say N. 2220 2221config TEST_HMM 2222 tristate "Test HMM (Heterogeneous Memory Management)" 2223 depends on TRANSPARENT_HUGEPAGE 2224 depends on DEVICE_PRIVATE 2225 select HMM_MIRROR 2226 select MMU_NOTIFIER 2227 help 2228 This is a pseudo device driver solely for testing HMM. 2229 Say M here if you want to build the HMM test module. 2230 Doing so will allow you to run tools/testing/selftest/vm/hmm-tests. 2231 2232 If unsure, say N. 2233 2234endif # RUNTIME_TESTING_MENU 2235 2236config MEMTEST 2237 bool "Memtest" 2238 ---help--- 2239 This option adds a kernel parameter 'memtest', which allows memtest 2240 to be set. 2241 memtest=0, mean disabled; -- default 2242 memtest=1, mean do 1 test pattern; 2243 ... 2244 memtest=17, mean do 17 test patterns. 2245 If you are unsure how to answer this question, answer N. 2246 2247 2248 2249config HYPERV_TESTING 2250 bool "Microsoft Hyper-V driver testing" 2251 default n 2252 depends on HYPERV && DEBUG_FS 2253 help 2254 Select this option to enable Hyper-V vmbus testing. 2255 2256endmenu # "Kernel Testing and Coverage" 2257 2258endmenu # Kernel hacking 2259